Online Submissions: http://www.journaltcm.

com
info@journaltcm.com

J Tradit Chin Med 2014 August 15; 34(4): 381-391

ISSN 0255-2922
2014 JTCM. All rights reserved.

SYSTEMATIC REVIEW

Efficacy of acupuncture on fibromyalgia syndrome: a Meta-analysis

Yang Bai, Yi Guo, Hong Wang, Bo Chen, Zhankui Wang, Yangyang Liu, Xue Zhao, Yinhong Li
aa
Yang Bai, Acupuncture and Moxibustion College, Tianjin
University of Traditional Chinese Medicine, Tianjin 300193,
China; Qibo Research Institute of Traditional Chinese Medicine, Qingyang Hospital of Traditional Chinese Medicine,
Gansu 745000, China
Yi Guo, Hong Wang, Bo Chen, Zhankui Wang, Yangyang
Liu, Xue Zhao, Yinhong Li, Acupuncture and Moxibustion
College, Tianjin University of Traditional Chinese Medicine,
Tianjin 300193, China
Supported by the State Key Program of National Natural
Science of China (No. 81330088)
Correspondence to: Prof. Yi Guo, Acupuncture and Moxibustion College, Tianjin University of Traditional Chinese
Medicine, Tianjin 300193, China. guoyi_168@163.com
Telephone: +86-22-59596221
Accepted: November 22, 2013

but no difference after 7 weeks of therapy. There

was no difference in the numerical rating scale in
weeks 3, 8 and 13. (b) Acupuncture versus drugs.
There were differences in the VAS after 20 days of
acupuncture and moxibustion treatment comparing with the drug amitriptyline, and after 4 weeks
of acupuncture and moxibustion treatment comparing with the drug fluoxetine and amitriptyline.
There were also differences in the number of tender points when comparing acupuncture with amitriptyline or fluoxetine. There was no difference in
total efficiency when comparing acupuncture with
amitriptyline after 4 weeks of treatment, but there
were differences between the two groups 45 days
after treatment. There were also differences in total
efficiency comparing acupuncture with fluoxetine,
and when comparing 4 weeks post-treatment of
acupuncture with a combination of amitriptyline,
oryzanol and vitamin B. (c) A comparison of acupuncture, drugs and exercise with drugs and exercise showed PPT differences in months 3 and 6.
There was no difference between the two comparison groups after follow-up visits in months 12 and
24.

Abstract
OBJECTIVE: To comprehensively evaluate the effectiveness of acupuncture as a treatment for fibromyalgia syndrome.
METHODS: Two review authors independently selected the trials for the Meta-analysis, assessed
their methodological quality and extracted relevant data. A quality assessment was conducted according to the Cochrane Review Handbook 5.0.
RevMan 5.0.20 software was used in the statistical
analysis.

CONCLUSION: Compared with sham acupuncture,

there was not enough evidence to prove the efficacy of acupuncture therapy for the treatment of fibromyalgia. Some evidence testified that the effectiveness of acupuncture therapy for fibromyalgia
was superior to drugs; however, the included trials
were not of high quality or had high bias risks. Acupuncture combined with drugs and exercise could
increase pain thresholds in the short term, but
there is a need for higher quality randomized controlled trials to further confirm this.

RESULTS: A total of 523 trials were reviewed and 9

trials were selected for Meta-analysis. (a) Compared
acupuncture with sham acupuncture, there was a
significant difference in the visual analogue scale,
but no difference in the pressure pain threshold.
Additionally, and there was a difference in the fibromyalgia impact questionnaire and the multidisciplinary pain inventory after 4 weeks of treatment,
JTCM | www. journaltcm. com

2014 JTCM. All rights reserved.

381

August 15, 2014 | Volume 34 | Issue 4 |

Bai Y et al. / Systematic Review

METHODS

Key words: Acupuncture; Fibromyalgia; Meta-analysis; Randomized controlled trial; Controlled clinical trial

Eligibility criteria
The research type: the chosen trials were either randomized controlled trials (RCT) or controlled clinical
trials (CCT). CCT did not comply strictly with the
random distribution method. For example, the distribution was in accordance with admission sequence or
other not-genuine randomized methods. The language
was limited to Chinese and English.
The research objects: there were no limits to research
subjects' age, gender, treatment courses, or source. The
definite diagnostic criteria were in line with FMS diagnostic criteria established by the American College of
Rheumatology (ACR) in 1990.21
The intervention types: the treatment group received
acupuncture therapy (no limits in needle type, needle
size and needle amounts, acupoint prescriptions, operating techniques, needle retention time, and course of
treatment); the control groups received sham acupuncture (nonpoints were stimulated and not stimulated on
the surface of the skin) or took Western Medicine (no
limits in type and dose). In addition, the treatment
group which received acupuncture combined with the
certain therapy and the control group with the same
certain therapy were both included. The acupuncture
therapy included filiform needle acupuncture, electro-acupuncture (EA), moxibustion, laser irradiation,
and point application.
The outcome indicators: the major outcome indicators
were a visual analogue scale (VAS), and a numerical rating scale (NRS). The minor outcome indicators were:
the number of tender points (TePsN), the pressure
pain threshold (PPT), the short-form health survey
(SF-36), the fibromyalgia impact questionnaire (FIQ),
the multidisciplinary pain inventory (MPI), and the total efficacy rate.

INTRODUCTION
Fibromyalgia syndrome (FMS) is a non-joint rheumatism, that clinically mainly manifests as diffused skeletal muscle pain and systemic symmetrical distributed
tender points,1 accompanied by symptoms such as fatigue, depression, anxiety, dipsomania, headaches, diffuse abdominal pain, and frequent micturition that severely affect the patient's quality of life.2,3 To date,
there are no domestic epidemiology statistics for this
disorder. The American Rheumatism Association
(ARA) states that FMS is the third most common rheumatic disease, after rheumatoid arthritis (RA) and osteoarthritis (OA).4 The incidence of FMS is approximately 2%-4% with a female to male gender ratio of
approximately 91.5-6 The predilection age focuses on
35-507 and the pathogenesis is not yet known.8-10 A
study indicated that kinship to patients with FMS
means a higher susceptibility, suggesting it is related to
both genes and environmental factors.11
The European League Against Rheumatism (EULAR)
currently regards the drug amitriptyline as the most effective for FMS treatment, but its side effects hinder its
use as a long-term therapeutic method. As an economical therapeutic method, however, acupuncture and
moxibustion have been used to treat pain syndrome for
more than 2000 years.12 Studies suggest that 60%-90%
of FMS patients use one or more complementary or alternative therapeutic methods;13,14 of these 22% try acupuncture and moxibustion therapy.15 FMS therapeutic
guidelines moderately recommend acupuncture and
moxibustion as they may improve FMS symptoms.16,17
A systematic review and Meta-analysis can insure the
quality of a specialized-recommendation therapeutic
schedule that has superior clinical directive significance. At present, China's clinical trials do not include
international systematic reviews15,18,19 on acupuncture
versus sham acupuncture interventions whose results
reveal no evidence verifying that acupuncture therapy
achieves better results than sham acupuncture. Similarly, domestic systematic reviews20 that demonstrate that
interventions achieve better results than amitriptyline
in the treatment of FMS are not included in international clinical trials.These reviews all had methodological limitations and were published prior to 2010.
Hence, the current study thoroughly researched the
randomized and quasi-randomized controlled trials
both in China and abroad. On the basis of systematic
reviews, outcome indicators were chosen that objectively reflected the clinical curative effect and presented
the overall evaluation for clinical efficacy of acupuncture versus placebo and Western Medicine as well as
acupuncture comprehensive therapy.
JTCM | www. journaltcm. com

Information sources
Information was sourced by electronic retrieval from
Chinese databases such as China National Knowledge
Infrastructure Database (1979-2012), China Science
and Technology Journal Database (1989-2012), Wanfang Database (1998-2012), and from the English databases PubMed (1966-2012), EMBASE (1980-2012)
and Cochrane Library (fourth issue, 2012). Data were
also manually retrieved by searching library back issues
and recently published literature not contained in the
above databases.
Search strategy
The Chinese search terms used were: 'acupuncture and
moxibustion', 'needling' and 'fibromyalgia'. The English search terms were: 'fibromyalgia', 'fibromyal*', and
'acupuncture'. The period searched was until March 1,
2012. The specific retrieval strategy was:
To locate FMS: #1 fibromyalgia [MeSH]; #2 fibromyal* [tw]; #3 OR/1-2.
382

August 15, 2014 | Volume 34 | Issue 4 |

Bai Y et al. / Systematic Review

To locate acupuncture interventions: #4 acupuncture

[MeSH]; #5 acupuncture therapy [MeSH]; #6 acupuncture points [MeSH]; #7 body acupuncture [tw]; #
8 electroacupuncture [MeSH]; #9 electro-acupuncture
[tw]; #10 electrical acupuncture [tw]; #11 ear acupuncture [MeSH]; #12 auricular acupuncture [tw]; #13
scalp acupuncture [tw]; #14 OR/4-13; #15 3 AND 14.

ate publication bias (over 10 trials at minimum).

Dealing with missing data: the trial author was contacted when the mean or standard deviation was absent. If
the data remained unavailable, the standard deviation
was estimated through standard error, P-value, t-value,
or the mean was replaced with the median if the original data was under a normal distribution.

Data extraction
Two evaluators independently reviewed each study title and abstract. After excluding studies that clearly did
not meet the inclusion criteria, the remaining trials
were read in full for further determination. The reviewers cross-checked the test results and differences in
opinions were resolved by discussion or by third party
arbitration.

RESULTS
Study selection
In total, 677 relevant articles were identified: 551 of
these met the inclusive criterion, and 126 duplicated articles were excluded. Articles from Chinese journals in
English, non-English language literature, systematic reviews, literature reviews, case reports, and specialists'
experience were excluded, leaving a total of 52 articles.
The full text of these was carefully reviewed with only
nine final articles being selected for Meta-analysis (reasons for exclusion: 10 case observations, 10 repetitive
articles, seven non-randomized controlled trials, 15
other intervention models, and one without original
data). Data collection process as shown in Figure 1.

Risk of bias in individual studies

Following the quality assessment standard recommended by the Cochrane Review Handbook 5.0,22 the bias
risk assessment tool involved six aspects: (a) random
distribution method; (b) allocative decision concealment; (c) whether the research objects, therapeutic
plan operators, and those measuring the results were
blinded; (d) result integrity; (e) presenting the study
findings selectively; and (f) other bias resources. Each
research result was examined based on the above six aspects and judged as "YES" (low-degree bias), "NO"
(high-degree bias) or "unclear" (lacking relative information or uncertain bias condition). Two evaluators
cross-verified the quality assessment results of the inclusive trials and differences in opinions were resolved by
discussion or by third party arbitration.

Study characteristics
Research type: of the nine final studies, six24-27,30,32 were
RCT and three28,29,31 were CCT.
Research object: of the nine articles, one29 did not mention the patient source, four27,28,30,31 dealt with outpatients and/or inpatients, and the remaining four24-26,32
studied recruited patients. All nine trials were in line
with the diagnostic criteria established by the ARA in
1990.
Research interventions: in the trial test groups, two
studies24,25 used EA, one31 used transcutaneous electrical
stimulation, one27 applied laser irradiation to acupuncture points, one32 employed acupuncture combined
with antidepressant drugs and exercise, and the remaining four treated with acupuncture. In the trial control
groups, three trials27,28,31 used amitriptyline, one29 used
fluoxetine, one30 used amitriptyline with the oryzanol
and vitamin B1, three24-26 used sham acupuncture, and
one32 used antidepressants with exercise (Table 1).
The selected acupuncture points: one trial25 chose the
points based on acupuncture and moxibustion literature, two trials27,28 adopted clinical experiences combined with the theory of Chinese medicine point selection, and the others employed Chinese medical theory
point selection (Table 2).

Summary measures and synthesis of results

Data Meta-analysis was conducted using RevMan
5.0.20 software,23 Each chosen study was tested for heterogeneity, and was considered heterogeneous if P<0.1
or I2>50%. A fixed effect model was used if no statistical heterogeneity existed in each study; if heterogeneity
existed, its origin was established. If clinical or methodological heterogeneity did not exist, a random effect
model was employed. Descriptive analysis was used if
distinct clinical heterogeneity existed in each study.
Weighted mean difference (WMD) was used for continuous variables, relative risk (RR) for categorical variables, and a 95% confidence interval (CI) signified every effect size, with P0.05 being judged to have a statistical significant meaning.
Risk of bias across studies and additional analyses
Subgroup analysis: to inspect the relationship between
acupuncture therapeutic efficacy and the therapeutic
course.
Sensitivity analysis: to check the stability of the results
and exclude lower-quality literature (unclear allocation
concealment) and trials with over 20% drop out rate.
Publication bias: applying RevMan software23 to evaluJTCM | www. journaltcm. com

Risk of bias within studies

Random allocation method: four trials24-26,32 used computers to generate the random allocation sequence,
one30 used a random number table to generate the random sequence, one used the ballots method,27 and the
others28,29,31 were randomized sequences according to
the date of attending the doctor.
383

August 15, 2014 | Volume 34 | Issue 4 |

Included

Eligibility

Screening

Identification

Bai Y et al. / Systematic Review

Records identified
through database
searching (n=677)

Additional records
identified through
other sources (n=0)

Records after duplicates

removed (n=551)

Articles assessed
for eligibility
(n=52)

Studies included
in qualitative synthesis (n=9)

Articles excluded (n=42)

case series (n=10)
multiple publication
(n=10)
Non-randomized
controlled trials (n=7)
Not meeting eligibility
criteria (n=15)
Not acquiring original
data (n=1)

Studies included
in quantitative
synthesis (Metaanalysis) (n=9)

Figure 1 Flow chart of report selection process

Allocation concealment: two trials24,26 used allocation

concealment and the others did not describe it.
Blinding method: two trials25,26 used blinding methods,
and the remainder did not describe whether they did.
Selective research report: one trial31 had a selective research report bias risk, the report of a further trial29 was
not clearly depicted, and the other trials had no selective report bias.
Other bias sources: only one trial26 stated that it had no
other bias risk; the others did not determine whether
there were other bias sources. The specific bias analysis
of each test is shown in Figures 2 and 3.

trast ordinary acupuncture and sham acupuncture

groups based on the difference in curative effect, using
the FIQ scale. After 4 weeks of treatment, the measured outcomes revealed that the two groups had statistically significant FIQ scale differences [WMD=7.40,
95% CI ( 13.60, 1.20)]. After 7 weeks of treatment, the results showed that the two groups had no
statistically significant FIQ score differences [WMD=
4.60, 95% CI (10.65, 1.45)].
MPI scale evaluation: one trial25 was included to contrast ordinary acupuncture and sham acupuncture
groups based on the difference in curative effect, using
the MPI scale. After 4 weeks of treatment, the measured outcomes revealed that the two groups had statistically significant MPI score differences [WMD=7.40,
95% CI ( 13.12, 1.68)]. After 7 weeks of treatment, the results showed that the two groups had no
statistically significant MPI score differences [WMD=
4.10, 95% CI (10.20, 2.00)].
NRS scale evaluation: two trials26a/b (a/b: one article includes two different trials) were included to contrast ordinary acupuncture and sham acupuncture groups
based on the difference in curative effect. An NRS
scale evaluation was employed in weeks 3, 8 and 13.
The combined results in week 3 revealed that the two
groups had no statistically significant NRS score difference [WMD= 1.06, 95% CI ( 10.41, 8.30),
Chi2=0.03, I2=0% ]. The combined results in week 8
showed the two groups had no statistically significant

Synthesis of results
Acupuncture vs sham acupuncture: evaluation of VAS
pain scale: One trial24 was included to contrast EA and
sham acupuncture groups based on the difference in
curative effect, using the VAS pain scale. After 3 weeks
of treatment, the measured outcomes showed that the
two groups had statistically significant VAS score differences [WMD=13.89, 95% CI (28.86, 0.92)].
PPT scale score: Martin DP et al 25 was included to
contrast EA and sham acupuncture groups based on
the difference in curative effect, using the PPT scale.
After 3 weeks of treatment, the measured outcomes
showed that the two groups had no statistically significant PPT score differences [WMD=0.78, 95% CI
(0.01, 1.55)].
FIQ scale evaluation: one trial25 was included to conJTCM | www. journaltcm. com

384

August 15, 2014 | Volume 34 | Issue 4 |

Bai Y et al. / Systematic Review

Table 1 Main study characteristics
Study

Intervention

Course of treatment

Main outcome

3 weeks (2 treatment/week)

- VAS
- PPT
- FIQ
- MPI

Experimental intervention
Electro-acupuncture (n=36)

Control intervention
Sham acupuncture (n=34)

Electro-acupuncture (n=25)

Sham acupuncture (n=25)

4, 28 weeks (1 treatment/2
to 4 days during 2 to 3
weeks)

Harria RE
et al 2005a26
Harria RE
et al 2005b26

Traditional Chinese
acupuncture (n=25)
Traditional Chinese
acupuncture (n=25)

Nontraditional site with

stimulation (n=28)
Nontraditional site with
no stimulation (n=27)

- 3 weeks: 1 treatment/week
- 3 weeks: 2 treatment/week
- 3 weeks: 3 treatment/week

- NRS (pain intensity)

- MFI (fatigue)
- SF-36 (PF)

Wang CM
200827

Amitriptyline (n=28)

20 days

- VAS

Guo AS
et al 200528

Traditional Chinese
acupuncture together with
acupoint laser irradiation
(n=28)
Traditional Chinese
acupuncture (n=19)

Amitriptyline (n=19)

4 weeks

Guo Y
et al 201029

Electro-acupuncture
together with TDP (n=36)

Fluoxetine (n=35)

4 weeks

Wang SP
et al 200230

Traditional Chinese
acupuncture (n=28)

4 weeks

Guo XJ
et al 2004a31
Guo XJ
et al 2004b31
Targino RA
et al 200832

Dermal neurological
electrical stimulation (n=22)
Chinese
electro-acupuncture (n=22)
Traditional Chinese
acupuncture together with
tricyclic antidepressants and
exercise (n=34)

Amitriptyline together
with Oryzanol and
Vitamin B1 (n=28)
Amitriptyline (n=22)

- VAS
- TePsN
- Total efficiency
- VAS
- TePsN
- Total efficiency
- MPQ (PRI, PPI)
- Total efficiency

Deluze C
et al 199224
Martin DP
et al 200625

45 days

- Total efficiency

12, 24, 48, 96 weeks (2

treatment/week)

- VAS
- TePsN
- PPT
- SF-36 (PF, RP, BP,
GH, VT, SF, RE, MH)

Amitriptyline (n=22)
tricyclic antidepressants
and exercise (n=24)

Notes: a/b: one article includes two different trials. VAS: visual analogue scale; PPT: pressure pain threshold; FIQ: fibromyalgia impact
questionnaire; MPI: multidisciplinary pain inventory; NRS: numerical rating scale; MFI: multi-dimensional fatigue inventory; SF-36:
short-form health survey (PF: physical functioning, RP: role physical, BP: bodily pain, GH: general health, VT: vitality, SF: social functioning, RE: role emotional, MH: mental health); TePsN: the number of tender points; TDP: specific electromagnetic spectrum treatment
device; MPQ: McGill pain questionnaire (PRI: pain rating index, PPI: present pain intensity).

NRS score differences [WMD=6.80, 95% CI (3.66,

17.25), Chi2=1.17, I2=15% ]. The combined results in
week 13 showed the two groups had no statistically significant NRS score differences [WMD=4.19, 95% CI
(6.86,15.24) Chi2= 0.75, I2=0%] (Figure 4).

(3.13, 1.81)] (Figure 5).

Evaluation of TePsN (the number of tender points)
Two included trials28,29 investigated TePsN, and a TePsN tender point count was conducted after 4 weeks of
treatment in both. One trial28 compared acupuncture
with amitriptyline, and the result proved that the two
groups had statistically significant TePsN differences
[WMD=4.00, 95% CI (6.73, 1.27)]. The other
trial29 contrasted the efficacy of acupuncture therapy
with fluoxetine, and the result revealed that the two
groups had statistically significant TePsN differences
[WMD=5.20, 95% CI (7.78, 2.62)].
Evaluation of total efficiency: five trials28-30,31a/b investigating total efficiency were included. Three trials28,31a/b
compared acupuncture with amitriptyline, one28 of
which measured the treatment outcome after 4 weeks
showing that the two groups had no statistically significant differences in the case of total efficacy [RR=1.38,
95% CI (1.00, 1.91)]. The other two trials31a/b adopted
the same indicators of efficacy and measured the out-

Acupuncture versus drugs

VAS pain scale: all three included acupuncture versus
drugs trials27-29 used the VAS pain scale. Two of these
compared the efficacy of acupuncture with amitriptyline: one27 of which measured the outcome on day 20
to reveal that the two groups had statistically significant VAS score differences [WMD= 2.27, 95% CI
( 3.05, 1.49)] and the other28 of which measured
the outcome in week 4 showing that the two groups
had statistically significant VAS score differences
[WMD= 17.10, 95% CI ( 23.93, 10.27)]. The
third trial29 compared the efficacy of acupuncture therapy with fluoxetine and measured the outcome in week
4 showing that the two groups had statistically significant VAS score difference [WMD= 2.47, 95% CI
JTCM | www. journaltcm. com

385

August 15, 2014 | Volume 34 | Issue 4 |

Bai Y et al. / Systematic Review

Table 2 Summary of treatment acupuncture points and rationale for selection of acupuncture points
Rationale for selection
Study
Acupuncture points
of acupuncture points
Martin DP
Hegu (LI 4), Zusanli (ST 36), Xingjian (LR 2),
Acupuncture literature
et al 200625
Sanyinjiao (SP 6), Shenmen (HT 7)

Adverse event
None (+)

Harria RE
et al 2005a/b26

Baihui (GV 20), Shangyang (LI 11), Hegu (LI 4),

Yanglingquan (GB 34), Zusanli (ST 36), Sanyinjiao (SP
6), Sanjian (LI 3)

TCM theory

Mild bruising and soreness

Mild vasovagal symptoms

Wang CM 200827

Ahshi-point

Not mentioned

Guo AS
et al 200528

Points of Governor Vessel, Urinary Bladder Meridian of

Foot-Taiyang, the first and second lateral line

TCM theory
Clinical experience
TCM theory
Clinical experience

Guo Y
et al 201029

Ahshi-point
Adjunct points: Pishu (BL 20), Weishu (BL 21), Zusanli
(ST 36), Hegu (LI 4), Jiexi (ST 41), Quchi (LI 11),
Sanyinjiao (SP 6), Guanyuan (CV 4), Shenshu (BL 23),
Shenmen (HT 7), Geshu (BL 17), Fengmen (BL 12),
Waiguan (TE 5), Taichong (LR 3)

TCM theory

Wang SP
et al 200230

Ahshi-point
Shaoshang (LU 11), Taiyuan (LU 9), Shangyang (LI 1),
Sanjian (LI 3)

TCM theory

Palpitation 0/6 (treatment

group/control group)
Mouth dryness 0/8
Dizziness 0/4
Perspiration 0/5
In appetence 0/4
Constipation 0/2
Not mentioned

Guo XJ
et al 2004a/b31

Main point: Fenchi (GB 20), Jianjing (GB 21), Xinshu

(BL 15), Dushu (BL 16), Geshu (BL 17), Zhibian (BL
54), Huantiao (GB 30), Huiyang (BL 35), Quchi (LI
11), Ququan (LR 8), Kufang (ST 15), Wuyi (ST 14)
Adjunct points: Taixi (KI 3), Shenmen (HT 7), Zusanli
(ST 36), Neiguan (PC 6)

TCM theory

Not mentioned

Targino RA
et al 200832

Hegu (LI 4), Zusanli (ST 36), Xingjian (LR 2),

Sanyinjiao (SP 6), Neiguan (PC 6), Yanglingquan (GB
34)

TCM theory

Not mentioned

Not mentioned

Notes: a/b: one article includes two different trials. TCM: Traditional Chinese Medicine.
Adequate sequence generation?
Allocation concealment?
Blinding?
Incomplete outcome data addresses?
Free of selective reporting?
Free of other bias?
0%

25%

50%

75%

100%

No (high risk of bias)

Unclear
Yes (low risk of bias)
Figure 2 Each methodological quality item presented as percentages across all included studies
Deluze C et al 199224

Guo XJ et al 2004a31

Guo AS et al 201029

Guo XJ et al 2004b31

Guo Y et al 200528

Harris RE et al 2005a26

Harris RE et al 2005b26

Martin DP et al 200625

Targino RA et al 200832

Wang SP et al 200230

Wang CM et al 200827

Adequate sequence generation?

Allocation concealment?
Blinding?
Incomplete outcome data addresses?
Free of selective reporting?
Free of other bias?
Figure 3 Each risk of bias domain for each included study
JTCM | www. journaltcm. com

386

August 15, 2014 | Volume 34 | Issue 4 |

Bai Y et al. / Systematic Review

come after 45 days of treatment. Their combined results showed that the two groups had statistically significant differences for total efficacy [RR=1.43, 95% CI
(1.16, 1.76), Chi2=0.03, I2=0% ]. One trial29 compared
the efficacy of acupuncture with fluoxetine after 4
weeks of treatment, and the results showed that the
two groups had statistically significant differences for
total efficacy [RR=1.60, 95% CI (1.18, 2.17)]. One trial30 compared the efficacy of acupuncture with amitriptyline, oryzanol and vitamin B after 4 weeks of treatment, and the results showed that the two groups had
a statistically significant difference [RR=1.50, 95% CI
(1.13, 1.99)] (Figure 6).

exercise. In this study, only the PPT scores were analyzed and the results showed a statistically significant
difference in both the first 3 months [WMD=0.69,
95% CI (0.38, 1.00)] and 6 months [WMD=0.57,
95% CI (0.25, 0.89)]. There was no statistically significant difference on follow-ups in months 12 and 24
(Figure 7).
Risk of bias across studies and additional analyses
Subgroup analysis. Comparing acupuncture with sham
acupuncture, there were significant effects on the MPI
and FIQ scores after 4 weeks of acupuncture treatment; however, there were no effects on MPI and FIQ
scores after 7 weeks of acupuncture treatment. Comparing acupuncture with Western Medicine (amitriptylin): after 4 weeks of acupuncture treatment there were
no effects on total efficacy; however, after 45 days,
there were significant effects.

Acupuncture, drugs and exercise vs Western

Medicine and exercise
Only one trial32 compared the efficacy of acupuncture,
antidepressants and exercise with antidepressants and

Figure 4 Meta-analysis of efficacy on acupuncture vs sham acupuncture by numerical rating scale

Figure 5 Meta-analysis of efficacy on acupuncture vs drugs by visual analogue scale

JTCM | www. journaltcm. com

387

August 15, 2014 | Volume 34 | Issue 4 |

Bai Y et al. / Systematic Review

Sensitivity Analysis. There was no significant heterogeneity in the NRS scale evaluation outcomes in weeks 3,
8 and 13. Because of the limited number of studies,
the potential sources of heterogeneity could not be assessed.

Publication bias. Because less than 10 studies were analyzed, a visual inspection of funnel plots for indicators
of publication bias was not undertaken.
Dealing with missing data. The VAS, TePsN, and
SF-36 data in one trial32 were all median, and it was

Figure 6 Meta-analysis of efficacy on acupuncture vs drugs by total efficacy rate

Figure 7 Meta-analysis of efficacy on acupuncture + drugs+ exercise vs drugs + exercise by pressure pain threshold
JTCM | www. journaltcm. com

388

August 15, 2014 | Volume 34 | Issue 4 |

Bai Y et al. / Systematic Review

not possible to establish the original data or to determine whether it was a normal distribution, thus the
median was not equivalent to the mean. Only the PPT
scores were analyzed in this case.

ority of amitriptyline. The present findings suggested

that acupuncture was superior to amitriptyline on days
20, 28 and 45, further certifying the superiority of acupuncture.
Four of the trials24,25,26a/b comparing acupuncture and
sham acupuncture had a high methodological quality;
however, three24,26a/b chose points away from the verum
acupoint for the sham acupuncture group and had a
negative result. The authors of this current paper consider this to be incorrect because (a) they may have
punctured other meridians or acupoints and thus generated certain therapeutic effects, (b) it is difficult to
find a true ineffective acupoint,37 (c) the width of the
meridians was unclear, thus whether the sham acupoint was on the verum meridian could not be established, and (d) it was not possible to establish if the
acupoints chosen in the trials were of superior efficiency because there is a lack of generally accepted superior
acupoint groups. In addition, the Meta-analysis revealed a general trend: there was statistical significance
in the VAS, FIQ, and MPI scales in weeks 3 and 4,
but this became negative after week 4. Therefore, there
is a need to establish if this difference was caused by
the effect of acupuncture or the scales on their own.
Aerobic exercise and anti-depressant drug recommended by American Pain Society was the evidence-based
A-level while acupuncture was C-level.38 However, in
one trial32 the effect of acupuncture was prominent
when the acupuncture, aerobic exercise and anti-depressant drug therapies were integrated, particularly in
the third month. Although blinding and allocation
concealment were not described in the trial, acupuncture would be the main therapy rather than other therapies with its increasing therapeutic effect if the integrated therapy was proven by more trials.39
In addition, the advantage of using acupuncture to
treat FMS was the low number of side effects. Three
RCTs25,26,29 assessed the adverse events of acupuncture
treatment. Acupuncture side effects were mild26 and infrequent25 compared with sham acupuncture, and
when compared with drug treatment, no severe adverse
acupuncture effects were noted.29 Applying acupuncture for the management of FMS might result in fewer
adverse effects than drug treatments.

DISCUSSION
Summary of evidence
In conclusion, there was not sufficient evidence to
prove that acupuncture had advantages in the treatment of FMS compared with sham acupuncture. However, for pain relief and reducing the number of tender
points, acupuncture proved superior to drugs. Because
of the high risk of bias from low-quality literature,
high quality RCT trials are needed to support the conclusion. Moreover, there is evidence that pain thresholds can increase with a combination of acupuncture,
Western Medicine and exercise in the short run (3-6
months), but there was no evidence of advantages in
the long run follow-up period (12-24 months).
Methodological strengths and limitations of included
trials
Comparing acupuncture with antidepressant was undertaken in five clinical trials in China but without foreign reports. Meta-analysis results have shown the superiority of acupuncture in the treatment of FMS. However, measurement bias was likely as there was no allocation concealment or blinding in the included research. Moreover, there was a baseline imbalance because of the selection bias led by quasi-randomized control trials in some included trials.28-30 It was not possible
to increase test efficiency by less combined data. The
blinding method was one of the 22 recommended
items in the CONSORT statement;33 however, it was
difficult to implement because of the strong operability
of acupuncture. This is also one of the reasons for the
low-quality of acupuncture literature. Therefore, the
present authors stress the importance of double-blinding for the trial subject and outcome operator and of
blind assessment in blind acupuncture trials.34,35 For instance, researchers could blindfold patients and prevent the trial subject from talking with others who
have accepted the acupuncture therapy to implement
blinding and reduce the risk of bias.25,26 With regard to
measuring the total effective rate, the trials applied varied measurement standards without clear sources leading to weak powers of test and specificity. Thus, the authenticity of the results needs to be proved by further
rigorous clinical trials.
Despite the methodology limitations, the superiority
of acupuncture in the treatment of FMS cannot be denied. A systematic review36 of a RCT trial comparing
amitriptyline and sham acupuncture suggested statistical significance after a 6-8 week period of taking drugs
in the aspect of relieving pain and fatigue; however, no
significance in week 12 revealed the short-lived superiJTCM | www. journaltcm. com

Limitations of the systematic review and

Meta-analysis
The Chinese literature included in this systematic review was of a generally low quality and there was a
high risk of bias from some of the quasi-randomized
control trials. Moreover, a small sample had a high bias
potential owing to the effect of certain elements. A focus on subgroup analysis, less combined data, and
weak test power were likely to generate false positive
conclusions. Therefore conclusions should be treated
cautiously. Moreover, there was a certain clinical heterogeneity because of differences in acupoint application, course of treatment, course of disease, and age, as
389

August 15, 2014 | Volume 34 | Issue 4 |

Bai Y et al. / Systematic Review

well as in EA and acupuncture. Having searched the related literature published in China and abroad, the
present authors still cannot eliminate the potential of
publication bias.

Conclusion and future research

Acupuncture has a good short-run therapeutic effect in
treating FMS that is maintained for 1-3 months, with
the first month having the best therapeutic effect.
Therefore, the treatment of a patient with FMS for 1
month with acupuncture in conjunction with an anti-depressant drug and exercise therapy is recommended as the most favorable therapeutic effect.
In future research, the gold standard for acupoint selection rather than personal experiences, as well rigorous
trials of equivalence or non-inferiority in comparison
with Western Medicine should be first sought. A further research direction is the integration of acupuncture and Western Medicine such as an optimal selection of Western Medicine, acupoint selection and treatment course. Appropriate sham acupuncture and
non-therapeutic or ineffective acupoints should be explored for placebos. Given that acupuncture works by
different acupoint groupings (or prescriptions) to treat
diseases, it should therefore be described as effective or
ineffective that certain groups of acupoints contrast certain intervention measures. Thus, the single word 'acupuncture' is not recommended in the treatment of
FMS.
Acupuncture is increasingly used as a traditional therapy in western countries to treat musculoskeletal disease.40 We suggest that the diagnostic criteria proposed
by the ACR in 201041 should be adopted in future research. Moreover, based on the CONSORT Statement, a rigorously designed multi-center RCT and a
large practical sample should be implemented to assess
the clinical therapeutic effect of FMS by acupuncture
therapy.

10

11
12

13

14

15

16

17
18

19

REFERENCES
1

5
6

20

Merskey H, Bogduk N. Classifications of chronic pain.

Description of chronic pain syndromes and definitions of
terms. Seattle: IASP Press, 1994: 45.
Phillip J, Lesley M, Leslie J, et al. Identifying the clinical
domains of fibromyalgia: contributions from clinician and
patient delphi exercises. Arthritis Rheum 2008; 59(7):
952-960.
Hauser W, Zimmer C, Felde E, Kollner V. What are the
key symptoms of fibromyalgia syndrome? Results of a survey of the German Fibromyalgia Association. Schmerz
2008; 22(2): 176-183.
Louis A, Healey MD. A description of rheumatology practice. The American rheumatism association committee on
rheumatologic practice. Arthritis Rheum 1977; 20(6):
1278-1281.
Chakrabarty S, Zoorob R. "Fibromyalgia". Am Fam Physician 2007; 76(2): 247-254.
Bartels EM, Dreyer L, Jacobsen S, Jespersen A, Bliddal

JTCM | www. journaltcm. com

21

22

23

24

25
390

H, Danneskiold-Sams B. "Fibromyalgia, diagnosis and

prevalence. Are gender differences explainable?". Ugeskr
Laeger 2009; 171 (49): 3588-3592.
Jain AK. Bruce M, Marjorie I, et al. Fibromyalgia syndrome: Canadian clinicalworking case definition, diagnostic and treatment protocols- a consensus document. J Musculoskel Pain 2003; 11(4): 103-107.
Alarcon GS, Bradley LA. Advances in the treatment of fibromyalgia: current status and future directions. Am J
Med Sci 1998; 315(6): 397-404.
Pongratz DE, Sievers M. Fibromyalgia-symptom or diagnosis: a definition of the position. Scand J Rheumatol Suppl 2000; 113(13): 3-7.
Senna ER, De Barros AL, Silva EO, et al. Prevalence of
rheumatic diseases in Brazil: a study using the COPCORD approach. J Rheumatol 2004; 31(3): 594-597.
Neumann L, Buskila D. Epidemiology of fibromyalgia.
Curr Pain Headache Rep 2003; 7(5): 362-368.
Dimmock S, Troughton PR, Bird HA. Factors predisposing to the resort of complementary therapies in patients
with fibromyalgia. Clin Rheumatol 1996; 15(5): 478-482.
Pioro-Boisset M, Esdaile JM, Fitzcharles MA. Alternative
medicine use in fibromyalgia syndrome. Arthritis Care Res
1996; 9(1): 13-17.
Bombardier CH, Buchwald D. Chronic fatigue, chronic
fatiguesyndrome,andfibromyalgia.Disabilityand health-care
use. Med Care 1996; 34(9): 924-930.
Martin-Sanchez E, Torralba E, Daz-Domnguez E,
Barriga A, Martin JL. Efficacy of acupuncture for thetreatment of fibromyalgia: systematic review and Meta-analysis
of randomized trials. Open Rheumatol J 2009; 3: 25-29.
Berman BM, Ezzo J, Hadhazy V, Swyers JP. Is acupuncture effective in the treatment of fibromyalgia? J Fam
Pract 1999; 48(3): 213-218.
Lee TL. Acupuncture and chronic pain management. Ann
Acad Med Singapore 2000; 29(1): 17-21.
Mayhew E, Ernst E. Acupuncture for fibromyalgia-a systematicreview of randomized clinical trials. Rheumatology
2007; 46(5): 801-804.
Langhorst J, Petra K, Frauke M, Dominik I, Winfried H.
Efficacy of acupuncture in fibromyalgia syndrome--a systematic review with a Meta-analysis of controlled clinical
trials. Rheumatology (Oxford) 2010; 49(4): 778-788.
Gao X, Du YH, Li B, et al. Systematic review of clinical
trials for acupuncture vs durg in the efficacy for fibromyalgia syndrome. Zhong Guo Teng Tong Yi Xue Za Zhi
2010; 16(2): 112-114.
Wolfe F, Smythe HA, Yunus MB, et al. The American college of rheumatology 1990 criteria for the classification of
fibromyalgia. Report of the Multicenter Criteria Committee. Arthritis Rheum 1990; 33(2): 160-172.
Li J, Li YP. Improving and devoloping cochrane systematic review. Chinese Zhong Guo Xun Zheng Yi Xue Za Zhi.
2008; 8(9): 742-743.
Review Manager (RevMan) [Computer program]. Version 5.0.20 Copenhagen: The Nordic Cochrane Centre,
The Cochrane Collaboration, 2008.
Deluze C, Bosia L, Zirbs A, et al. Electroacupuncture in
fibromyalgia: results of a controlled trial. BMJ 1992; 305
(6864): 1249-1952.
Martin DP, Christopher DS, Brent AW, et al. ImproveAugust 15, 2014 | Volume 34 | Issue 4 |

Bai Y et al. / Systematic Review

26

27

28

29

30

31

32

ment in fibromyalgia symptoms with acupuncture: results

of a randomized controlled trial. Mayo Clin Proc 2006; 81
(6): 749-757.
Harris RE, Tian XM, Williams DA, et al. Treatment of fibromyalgia with formula acupuncture: investigation of
needle placement, needle stimulation, and treatment frequency. J Altern Complement Med 2005; 11(4): 663-671.
Wang CM. Clinical treatment of acupuncture and semiconductor laser irradiation on fibromyalgia symptoms 28
cases. Shi Yong Zhong Yi Nei Ke Xue Za Zhi, 2008; 22
(1): 58.
Guo Y, Sun YZ. Clinical study on treatment of fibromyalgia syndrome with penetration needling at the back.
Zhong Guo Zhen Jiu 2005; 25(2): 98-100.
Gou AS, Li AH. Clinical observation on therapeutic effect
of elec-acupuncture combining TDP irradiation on fibromyalgia syndrome 36 cases. Jiao Tong Yi Xue. 2010; 24
(4): 410-411.
Wang SP, Wang XF, Zhang DX, et al. Clinical observation on therapeutic effect of acupuncture treatment based
on syndrome differentiation of meridians on fibromyalgia
syndrome. Zhong Guo Zhen Jiu 2002; 22(12): 807-809.
Guo XJ, Jia J. Comparison of therapeutic effects on fibromyalgia syndrome between dermal-neurological electric
stimulation and electric acupuncture. Zhong Hua Wu Li
Yi Xue Yu Kang Fu Za Zhi 2004; 26(2): 126-127.
Targino RA, Alves R, Imamura M, et al. A randomized
controlled trial of acupuncture added to usual treatment
for fibromyalgia. J Rehabil Med 2008; 40(7): 582-588.

JTCM | www. journaltcm. com

33

34

35

36

37
38

39
40

41

391

Begg C, Cho M, Eastwood S, et al. Improving the quality

of reporting of randomized controlled trials. The CONSORT state-ment. JAMA 1996; 276(8): 637-639.
Hopton AK, Macpherson H. Assessing blinding in randomised controlled trials of acupuncture: challenges and
recommendations. Chin J Integr Med 2011; 17(3):
173-176.
White AR, Filshie J, Cummings TM. Clinical trials of
acupuncture: consensus recommendations for optimal
treatment, sham controls and blinding. Complement Ther
Med 2001; 9(4): 237-245.
Nishishinya B, Urrutia1 G, Walitt B, et al. Amitriptyline
in the treatment of fibromyalgia: a systematic review of its
efficacy. Rheumatology 2008; 47(12): 1741-1746.
Lao LX. Acupuncture clinical studies and evidence-based
medicine acupuncture research 2008; 33(1): 53-58.
Huser W, Thieme K, Turk DC. Guidelines on the management of fibromyalgia syndrome a systematic review.
Eur J Pain 2010; 14(1): 5-10.
NIH Consensus Conference. Acupuncture. JAMA 1998;
280: 1518-1524.
Green S, Barnsley L, Hall S, et al. Acupuncture for lateral
elbow pain. Cochrane Database Syst Rev 2002; 2002,(1):
CD003527.
Aletaha D, Neogi T, Silman AJ, et al. 2010 Rheumatoid
arthritis classification criteria: an American College of
Rheumatology/Europe-an LeagueAgainst Rheumatism collaborative initiative. Arthritis Rheum 2010; 62(9):
2569-2581.

August 15, 2014 | Volume 34 | Issue 4 |